Experimentation is a method by which scientists examine natural phenomena in the hope of gaining new knowledge. A good experiment follows a logical design to isolate and test a specific variable that is precisely defined. By learning the fundamental principles behind experimental design, you will be able to apply these principles to your own experiments. No matter the scope, all good experiments operate according to the logical and deductive principles of the scientific method, from fifth grade potato clock projects to advanced Higgs Boson research.
Step
Method 1 of 2: Designing Scientific Experiments
Step 1. Choose a specific topic
Experiments whose results lead to a change in scientific mindset are very, very rare. Most experiments answer certain small questions. Scientific knowledge is built from the accumulated data of many experiments. Choose a topic or unanswered question that is small in scope and easy to test.
- For example, if you want to experiment with agricultural fertilizers, don't try to answer the question, "What kind of fertilizer is best for growing crops?" There are many different types of fertilizers and many different types of plants in the world - one experiment cannot yield universal conclusions for both. A better question for designing the experiment would be "What concentration of nitrogen in the fertilizer produced the largest corn crop?"
- Modern scientific knowledge is very, very broad. If you intend to do scientific research, research your topic at length before you start designing your experiment. Have any previous experiments answered questions that were the subject of your experiment's learning? If so, is there a way to adapt your topic to answer questions that have not been answered by existing experiments?
Step 2. Isolate your variables
Good science experiments test specific, measurable parameters called variable.
In general terms, a scientist conducts an experiment for the value of the variable he is testing. One vital thing when conducting experiments is adjusting only the specific variable you are testing (and no other variables).
For example, in our fertilizer experiment example, our scientist will plant several large corn plants in soil that is fertilized with different nitrogen concentrations. It will give each plant the required amount of fertilizer exactly same. He will ensure that the chemical composition of the fertilizers used does not differ other than the nitrogen concentration - for example, he will not use fertilizers with higher magnesium concentrations for any of his maize crops. He will also plant the same number and species of corn plants at the same time and on the same soil type in each of his experimental replicas.
Step 3. Create a hypothesis
The hypothesis is a prediction of the experimental results. This should not be just guesswork - a good hypothesis is informed from the research you have done when choosing an experiment topic. Base your hypothesis on the results of similar experiments conducted by other colleagues in your field, if you are solving a problem that has not been studied in depth, based on any combination of literary research and recorded observations you can find. Remember that even if you do your best research, your hypothesis may be proven wrong - in this case, you are still expanding your knowledge by proving your prediction "not" right.
Typically, hypotheses are expressed as quantitative declarative sentences. A hypothesis also uses the way experimental parameters are measured. A good hypothesis for our fertilizer example is: "A corn plant fed one pound of nitrogen per bushel will produce a greater mass of yield than an equivalent corn crop grown with a different nitrogen supplement
Step 4. Plan your data collection
Know in advance " when " you will collect data, and " what type of " data you will collect. Measure this data at predetermined times, or in other cases, at regular intervals. In our fertilizer experiment, for example, we will measure the weight of our corn plant d(in kilograms) after a period of growth. We will compare this with the nitrogen content of the fertilizer applied to each plant. In other experiments (such as those that will measure changes in a variable over time), it is necessary to collect data at regular intervals.
- Creating a data table ahead of time is a good idea - you simply enter your data values into the table as you record it.
- Know the difference between dependent and independent variables. The independent variable is the variable that you change and the dependent variable is the one that is affected by the independent variable. In our example, "nitrogen content" is the "independent" variable, and "yield (in kg)" is the "dependent" variable. The base table will have columns for both variables as they change over time.
Step 5. Conduct your experiment methodically
Run your experiment, test for your variables. This almost always requires you to experiment repeatedly for some variable values. In our fertilizer example, we will plant several identical maize crops and apply a fertilizer containing varying amounts of nitrogen. Generally, the more extensive data you get, the better. Record as much data as possible.
- A good experimental design incorporates what is known as control. One of your replica experiments should "not" include the variable you are testing at all. In our fertilizer example, we will include one corn plant that receives fertilizer without nitrogen in it. This will be our control - will be the baseline against which we will measure the growth of other corn crops.
- Observe any and all safety related substances or processes in your experiment.
Step 6. Collect your data
Record data directly on the table, if possible - this will prevent you from having to re-enter and merge data later. Learn how to assess the extraneous in your data.
It's always a good idea to depict your data as visually as possible. Create data points on the chart and express trends with the most appropriate line or curve. This will help you (and anyone else viewing this graph) visualize patterns in the data. For most basic experiments, the independent variable is plotted on the horizontal x-axis and the variable alternating on the vertical y-axis
Step 7. Analyze your data and draw conclusions
Is your hypothesis correct? Are there any observable trends in the data? Did you find any unexpected data? Do you have unanswered questions that might form the basis for future experiments? Try answering these questions while you assess the results. If your data doesn't provide a definite "yes" or "no" hypothesis, consider conducting additional experimental trials and gathering more data.
To share your results, write a comprehensive scientific paper. Knowing how to write scientific papers is a rewarding skill - the results of recent research must be written and published in a certain format
Method 2 of 2: Running Example Experiments
Step 1. Choose a topic and define your variables
For the reason of this example, we will have a simple and small experiment. In our experiment, we will examine the effect of different aerosol fuels on the firing range of the potato gun.
- In this case, the type of aerosol fuel we are using is the "independent variable" (the variable we will change), where the bullet distance is the "dependent variable".
- Something to consider in this experiment - is there a way to make sure each potato bullet weighs the same? Is there a way to use the same amount of fuel for each shot? Both of these can affect the firing range of the gun. Measure the weight of each bullet first and use the same amount of aerosol spray for each shot.
Step 2. Create a hypothesis
If we test hair spray, cooking spray, and spray paint, let's say hair spary contains aerosol fuel with a butane content greater than other sprays. Since we know that butane is flammable, we can hypothesize that the hair spray will produce more thrust when ignited, shooting a potato bullet farther away. We will write a hypothesis: "The higher butane content of the aerosol fuel in the hair spray, on average, will produce a longer firing range when firing potato bullets weighing between 250-300 grams."
Step 3. Set up your previous data collection
In our experiment, we will test each aerosol fuel 10 times and calculate the average yield. We will also test an aerosol fuel that does not contain butane as an experimental control. To prepare, we will assemble our potato cannon, test it to make sure it works, purchase an aerosol spray and then cut and weigh our potato bullet.
-
We will also create a data table first. We will have five vertical columns:
- The leftmost column will be labeled "Test #". Cells in this column will contain the numbers 1-10, indicating each firing attempt.
- The next four columns will be labeled with the name of the aerosol spray we used in the experiment. Ten cells under each column header that will contain the distance (in meters) of each firing attempt.
- Under each of the four columns for fuel, leave space to write the average value for each distance.
Step 4. Do the experiment
We will use each aerosol spray to fire ten bullets, using the same amount of aerosol to fire each bullet. After each shot, we'll use a tape measure to measure the distance between each bullet. Record this data in a data table.
Like many experiments, our experiment has some safety issues that we must observe. The aerosol fuel we use is flammable - we must close the cover of the potato gun shooter well and wear thick gloves when igniting the fuel. To avoid accidental injury from bullets, we must also ensure that we (or other bystanders) stand by the gun while firing - not in front of or behind it
Step 5. Analyze the data
Say, we find that, on average, hair spray shoots potatoes the furthest, but cooking spray is more consistent. We can visualize this data. A good way to illustrate the average distance per spray is a bar graph, where a scatter plot is a great way to show variations in the firing distance of each fuel.
Step 6. Draw your conclusions
View the results of your experiments. Based on our data, we can say with confidence that our hypothesis is correct. We can also say that we found something we didn't predict - that cooking spray produced the most consistent results. We can report any issues or messes we find - let's say paint from spray paint builds up in a potato cannon firing chamber, making repeated firing difficult. Lastly, we can suggest areas for further research - for example, maybe with more fuel, we can get more distances.
We could even share our results with the world in the form of scientific papers - since the subject of our experiments, it may be more appropriate to present this information in the form of a trifold of science exhibitions
Tips
- Have fun and stay safe.
- Science is about asking big questions. Don't be afraid to pick a topic you haven't seen before.
Warning
- Wear eye protection.
- If anything gets into your eyes, rinse thoroughly for at least 5 minutes.
- Do not place food or drink near your workplace.
- Wash hands before and after the experiment.
- When using sharp knives, hazardous chemicals, or hot fires, make sure an adult is watching you.
- Wear rubber gloves when handling chemicals.
- Tie hair back.
